By Julie Sevrens Lyons
In what is being heralded as a potential breakthrough for patients with spinal cord injuries, a quadriplegic was able to open e-mail, change television stations and turn on lights with just his mind -- and a sophisticated brain sensor that translated his thoughts into actions.
Matthew Nagle, 25, was injured in a knife attack in 2001 and cannot use his arms or legs. But for the nine months he had a sensor implanted in his brain, he was able to play video games, draw rudimentary figures with a computer paint program and grab a piece of candy with a prosthetic hand.
In a complementary study, monkeys at Stanford University used the same technology to control computer cursors, accomplishing the task four times faster than in previous primate experiments. This, researchers said, puts the technology within the realm of practicality for patients, although it will take years to move it from the lab to the marketplace.
The dual findings, which appear today in the journal Nature, are raising hopes that brain power can be harnessed and put to work in some patients with physical impairments, allowing them to do mundane tasks that healthy people take for granted.
While the ultimate goal is to enable paralyzed people to move their own limbs again, experts say the independence gained by controlling a computer or a television remote control is not to be taken lightly.
``When you think about the spinal cord injury field, it was only a short time ago there wasn't any credible research, only a short time ago when there wasn't any hope,'' said Susan Howley, director of research for the Christopher Reeve Foundation, which helped fund the Stanford experiment.
``We don't necessarily have to think about enormous leaps in order to improve the independence and quality of life for people living day to day with paralysis.''
Researchers said they were especially excited by the fact that the patient was still able to activate the part of the brain that controls movement, called the motor cortex, three years after being paralyzed.
They said the technology could ultimately benefit thousands of people with damage to their spinal cords, muscles or nerves. This includes not only paraplegics and quadriplegics, but also survivors of severe strokes and some patients with muscular dystrophy.
``This is a great first step, but there's still a lot to be learned,'' stressed Dr. Leigh Hochberg, a neurologist at Massachusetts General Hospital and Spaulding Rehabilitation Hospital and lead author of the study on humans.
Among the issues: how to extend the life of the tiny brain sensors, which begin to decline after a few months of use. Given that the sensors have to be surgically implanted into the brain, repairing or changing a malfunctioning device is no small task.
Still, patient advocates are invigorated by the notion that something that sounds like science fiction could be coming to fruition.
In ongoing experiments, patients have undergone surgery to have a sensor about the size of a baby aspirin inserted into the area of the brain responsible for movement. The sensor, made up of 96 microelectrodes, records electrical signals from the brain, which are then fed into a device and interpreted using special software.
For Nagle, who is now living at a rehabilitation center in Massachusetts, it didn't take long to master a computer cursor by willing it to move with his thoughts.
``No one knew whether this would work with humans or how long it would take to master,'' he told the Mercury News in a written statement. ``After the first couple days, I was able to move a computer mouse, just by thinking. And then we went on from there: I could turn on the TV, open and close lights and move a mechanical hand.''
It was, he said, ``an incredible journey.''
The accomplishments represent years of work in many Silicon Valley fields, said Krishna Shenoy, an assistant professor of electrical engineering at Stanford and one of the authors of the study on monkeys and brain signals.
``What we do is implant silicon-based electrode arrays made with a manufacturing process extremely similar to what Intel and all the other `fabs' in the valley use,'' he explained.
``We process brain signals using algorithms similar to those used by Cisco and Lockheed. We also then use low-power circuitry, which has everything to do with what mobile telecommunications firms in the area are using.''
Although the equipment now requires patients or primates to be hooked up to a computer system by a bulky cable that extends from their scalps, ``ultimately the way these systems have to be built is wireless,'' Shenoy said.
Both studies used the BrainGate Neural Interface System, made by Cyberkinetics in Foxborough, Mass.
The system isn't perfect. For instance, the sensor implanted in Nagle worked with only 75 percent to 85 percent accuracy as he carried out actions.
But the system may prove superior to technologies that are further along in the pipeline, Shenoy said. Voice-recognition systems that guide prosthetic devices, for example, aren't useful for stroke patients who have difficulty speaking clearly. Eye-operated equipment can feel unnatural and uncomfortable.
The Stanford researchers found that a brain-computer interface could be operated at about four times the speed of earlier experiments, making it for the first time fast enough to be practical for patients.
Put into perspective, users could tap their brain signals to type about 15 words per minute, rather than just three or four.
``The dream for the research is to one day hopefully be able to connect brain to limb,'' said Massachusetts General's Hochberg. ``We still have a lot to learn to get there, but this is exciting.''